This Opinion article is by Jay Borkakoti, Director of Home Insurance, UK and Ireland for LexisNexis Risk Solutions, and it takes a look at how new tech is set to disrupt the staid little niche of water and flood claims. It keeps on raining/freezing in Britain, so the problem isn’t going to go away, but there are some new ways of preventing the very worst damage from happening, and dealing with the aftermath, which can save both the industry and consumers a lot of time and money.
The ‘Beast from the East’ and storm Emma has caused a surge in escape of water claims (EOW) over recent weeks and senior figures in the claims sector have called on the insurance industry to unite to promote preventive measures. But how realistic is to suggest these claims can be prevented and what can the industry do to price more accurately for escape of water and mitigate the losses incurred?
The volume of escape of water claims certainly shows no signs of abating. These claims are widely accepted as the Achilles heel of the home insurance industry. Latest ABI data shows that, in the first nine months of 2017, domestic escape of water claims cost £483 million, rising 1% on the same period in 2016. What is really concerning is that this is up 24% on the first nine months of 2014.
Pricing for EOW is a key challenge. When you look at how home insurers assess risk, there are a whole host of factors to consider. Security of terraced homes or flats versus isolated detached homes, proximity to waterways with the potential to flood the property, the type of doors, windows and locks used, how often the home is left unoccupied, are some of those factors to name just a few. However, a leaky washing machine or burst frozen pipe can affect any home and a dodgy plumbing job during a house restoration project can happen anywhere (as I know from personal experience.) Geographic or property details don’t appear to have the same predictive power on EOW as they do for other perils.
This inability to pinpoint root causes means there is currently no “go-to” dataset that provides the level of detail required to price for escape of water losses.
Making it personal
Adding a further layer of complication, water damage can result from human actions, whether that is substandard DIY work, a bath allowed to overrun or an accidental knock which damages an exposed radiator pipe. And of course, some leaks become apparent immediately, while others could remain hidden and undetected causing significant damage.
This ‘human factor’ is reason enough to place personal data at the top of the agenda for helping insurers assess risk. The lack of ability to pinpoint root causes of the peril suggest to me that the best course of action is to pool market-wide claims data on EOW to create an industry score to help plug the gap in knowledge. This would clearly be in the interests of both the home-dweller and the insurance provider as it would create greater clarity around the key drivers of the increase in EOW claims and should mean fairer and more accurate premiums overall.
Creating the perfect score
Scoring is very effective, by creating a consolidated view of an individual’s risk and their likelihood to claim that can be delivered at speed, and at point of quote. It starts with past claims data contributed by the market. This is supplemented with public records data such as CCJs, the edited Electoral Roll, insolvencies data and potentially policy data as well. In the case of EOW specifically, there might also be some property attributes which are predictive, such as number of bathrooms, central heating type or age of property.
We can then bring in the ability to link different and disparate records for an individual customer to provide a singular view. For example, insurance providers might want to know quickly whether a home insurance customer is, or was ever, a motor insurance customer. By applying increasingly sophisticated analytics models combined with big data processing power, we can create a risk score that accurately predicts future loss.
This is quite different from the types of scoring approaches prevalent in the market which tend to be based on a customer’s payment history of financial products and history of financial default. Whilst these have shown a level of predictiveness on motor insurance, they don’t work so well on home. This is because they aren’t modelled on (tailored to) home insurance claims data, which should yield a stronger outcome.
The first step on this road is creating a contributory database of claims information. Contributory claims databases offer unique and invaluable insights into the market as a whole as well as individual policyholders and applicants.
The motor insurance industry has benefitted hugely from the implementation of contributory databases. Sharing No Claims Discount data has allowed motor insurance providers to price more accurately and identify potentially fraudulent applicants, whilst smoothing the customer journey and cutting operational costs. The sharing of claims data has also helped insurance providers with risk assessment and fraud prevention.
Overcoming the IoT hurdles
What about the potential of IoT (Internet of Things) in helping to detect, prevent and price for escape of water claims? I believe we are a few years away yet from sensors smart enough to fully prevent EOW claims, but there is a huge opportunity for enterprising insurers to tap into connected home technology to help mitigate such events and help price for future losses.
It is true that further take up of IoT should start to create a new eco-system of data and a potential new way of pricing in the next few years, but this is still the future. There are several challenges that need to be surmounted before smart home technology can really help reduce losses and improve pricing accuracy.
The first hurdle to overcome is consumer adoption. From our research, most consumers with smart home tech (30%) invested in the product for the convenience it offered – they see little benefit in the technology from a risk mitigation perspective. As such, the take up of such devices specifically to mitigate risk is low and the value of the data from these devices is pretty limited until consumer adoption picks up.
Although this could take years we can make some assumptions about the risk profile of someone with a smart home device versus someone without, but we do not know how much at this stage. To help drive adoption, insurers could consider discounts to those customers with smart home devices or pay for the device up front with the expectation of savings from lower or no claims costs. It will be a punt but this could provide the initial insights needed to promote further investment and consumer engagement.
The evolution of the smart home device market
The ultimate goal is to rate on data collected using smart devices. This will require a huge amount of data and processing power to harvest this from many different devices, normalise it and deliver it back to the market as a score – taking a similar approach to the telematics market on motor. This could well create new rating factors for insurance which may be complementary to how the sector prices today or completely different. For example, we could begin to understand the risk differential between someone who keeps their heating on all winter to someone that doesn’t.
The evolution of the smart home device market is key – these devices need to be cheaper to drive consumer adoption and they need to deliver the right quality of data to have value to the insurance sector and their customers in pricing home insurance and mitigating the risk of perils such as EOW.
Clearly the brand interaction opportunities for insurers are massive, allowing engagement with customers throughout the year rather than just at renewal time, but first the hurdle of the cost vs benefit to improve the quality of devices and adoption must be overcome.